1. Field of the Invention
The present invention relates to a text editing assistor, and more particularly to a text editing assistor which displays a prospective character sequence based on a character sequence which is input by a user.
2. Description of the Background Art
The rapid prevalence of cellular phone-based e-mail services in recent years has resulted in much increased opportunities for users to input text via key operations. Under such circumstances, various proposals have been made for realizing text editing assistors which enhance the efficiency and ease of text input operations performed by the user. Behind such needs is the fact that written Japanese consists of different sets of alphabets (hiragana, katakana, etc.), with which further types of characters (Chinese characters, or the English alphabet) may be mixedly used. Furthermore, when a Japanese sentence is typed on a keyboard (as in the case of a computer), it is commonplace to employ the English alphabet system (known as Romaji) to first phonetically type up the Japanese sentence as heard to obtain an input character sequence spelt in hiragana, and then convert portions of the input character sequence to other types of characters such as katakana or Chinese characters. Throughout the present specification, such conversion will be referred to as “finalization”, and the entire sequence obtained after the finalization will be referred to as a “final character sequence”. Note, however, that a final character sequence is not irrevocably “final”, but maybe subjected to further editing by the user, e.g., deletion and/or addition.
A typical example of conventional techniques for enhancing the efficiency and ease of text input operations is presenting prospective character sequences based on a character sequence which is input by a user. For example, for a given character sequence which is input by the user (composed of hiragana, katakana, or the English alphabet (Romaji)), a text editing assistor may present to the user a list of “conversion candidates” (i.e., words or fractions of words as spelt in Chinese characters), the list being presented in an estimated order of frequency. On the other hand, for a given character sequence which is input by the user (e.g., a character sequence “oha”), a text editing assistor may predict the rest of the word and present the result of the prediction as a “prediction candidate” (e.g., “ohayo”, or ‘good morning’). Hereinafter, more specific instances of such techniques will be described.
One specific example is a technique of presenting to the user a curtailed list of candidates, from which unlikely or impossible candidates are omitted. In this technique, automatic predictions are made in the middle of the user's inputting operation of a character sequence. Then, based on consistency with any previously-finalized sequence of Chinese characters, hiragana, and/or katakana, or user instructions as to which conversion candidates are adopted or dismissed, a curtailed list of candidates from which unnecessary prospective character sequences are omitted is presented to the user. Thus, the burden of the user's input operations is reduced.
Another specific example is a technique in which prospective character sequences to be presented to the user are “learned” based on input character sequences and finalized character sequences (final character sequences). According to this technique, a prospective character sequence which is selected as a final character sequence for a given input character sequence will be presented as a most prospective character sequence the next time the same character sequence is input. A further modification to a text editing assistor having such a learning function is a technique in which, if a once-finalized character sequence is corrected, the correction is reflected in the learning. This technique works in such a manner that, if a user has inadvertently selected the wrong final character sequence, this wrong final character sequence is prevented from being presented as the most prospective character sequence.
The purpose of using a text editing assistor, or the environment in which it is used, may vary. For example, the same user may use a text editing assistor to write a text for business purposes or use it in a more private situation. Since the frequency which a given character sequence is used varies depending on the purpose and/or environment associated with the text editing assistor, the character sequence to be presented as the most prospective character sequence should also vary depending on the purpose and/or environment associated with the text editing assistor. For example, a sentence like “to whom it may concern” is often used in business situations, and therefore, it would be convenient if such a sentence is presented as a prediction candidate when the user has input a character sequence “to”. On the other hand, such a sentence is less often used in private situations, so that it may detract from the user's convenience if the above sentence is presented as a prediction candidate in private situations.
As described above, prospective character sequence which are presented on a conventional text editing assistor do not reflect the purpose or state for/in which the text editing assistor is used. In other words, a conventional text editing assistor will always present the same prospective character sequence in response to a given input sequence, irrespective of the purpose or state. Therefore, depending on the purpose for which the text editing assistor is used, the function of presenting prospective character sequences may not provide much convenience to the user.